The overall goal of this revised application is to determine whether HER-2 and/or p53 abnormalities in primary breast cancer tissues predict whether women with node positive breast cancer benefit from either doxorubicin dose escalation or addition of paclitaxel in the adjuvant setting. These studies will be performed using tissues that were prospectively collected from patients who participated in an NCI-supported, multi-institutional U.S. Inter-Cooperative Group clinical trial led by the Cancer and Leukemia Group B (CALGB 9344/INT0148). In this clinical trial, node positive patients with newly diagnosed breast cancer received adjuvant chemotherapy consisting of doxorubicin (AdriamycinTM) and cyclophosphamide for four cycles (ACX4) in which the doxorubicin was administered at 60, 75, or 90 mg/m2 in a randomly assigned fashion. After ACX4, patients were also randomly assigned to receive paclitaxel (TaxolTM) for four cycles or no further therapy. The preliminary results of this clinical trial suggest that for all patients, there has been little or no added benefit for doxorubicin dose escalation, while a small but statistically significant improvement has been observed for those patients who received paclitaxel. Both of these strategies are associated with increased toxicities compared to standard dose AC alone. It would be of substantial clinical importance to determine factors that identify subgroups of patients that might be more likely to benefit from either doxorubicin dose escalation or addition of paclitaxel. The randomized nature of the 9344 trial design permits direct evaluation for interactions between these therapeutic strategies and potential predictive markers for benefit from doxorubicin dose escalation or addition of paclitaxel. The major hypotheses to be tested in this proposal are that abnormalities (amplification, overexpression) in two oncogenes, HER-2 and p53, which are common in breast cancer, might serve as such predictive markers. Preliminary data generated by the investigators, and others, strongly support the hypothesis that abnormalities in these two oncogenes may lead to enhanced benefit from doxorubicin dose escalation and addition of paclitaxel. In Specific Aim 1, we will determine if HER-2 amplification (as determined by fluorescent in situ hybridization, FISH) and/or overexpression (as determined by immunohistochemistry, IHC) predicts for benefit from either doxorubicin dose escalation or addition of paclitaxel. In Specific Aim 2, we will similarly determine whether p53 abnormalities, as determined by immunohistochemistry, predicts for benefit from either of these two therapeutic strategies. In conclusion, these studies should enable us to determine whether or not HER-2 and/or p53 are important molecular predictors of added benefit from doxorubicin dose escalation and/or addition of paclitaxel in node positive patient with newly diagnosed breast cancer. [unreadable] [unreadable]